| { |
| "best_global_step": 30, |
| "best_metric": 0.5807782411575317, |
| "best_model_checkpoint": "bert_base_rand_100_v1_mrpc/checkpoint-30", |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 105, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.7227780818939209, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6241, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6936274509803921, |
| "eval_combined_score": 0.7513938781619518, |
| "eval_f1": 0.8091603053435115, |
| "eval_loss": 0.5958898067474365, |
| "eval_runtime": 0.2695, |
| "eval_samples_per_second": 1513.971, |
| "eval_steps_per_second": 7.421, |
| "step": 15 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.493427038192749, |
| "learning_rate": 4.8e-05, |
| "loss": 0.5778, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.7107843137254902, |
| "eval_combined_score": 0.7570588235294118, |
| "eval_f1": 0.8033333333333333, |
| "eval_loss": 0.5807782411575317, |
| "eval_runtime": 0.2883, |
| "eval_samples_per_second": 1415.329, |
| "eval_steps_per_second": 6.938, |
| "step": 30 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.3310613632202148, |
| "learning_rate": 4.7e-05, |
| "loss": 0.5155, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.7083333333333334, |
| "eval_combined_score": 0.7586610486891385, |
| "eval_f1": 0.8089887640449438, |
| "eval_loss": 0.5847347974777222, |
| "eval_runtime": 0.2696, |
| "eval_samples_per_second": 1513.304, |
| "eval_steps_per_second": 7.418, |
| "step": 45 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 5.041624546051025, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.3911, |
| "step": 60 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.696078431372549, |
| "eval_combined_score": 0.7443602859203882, |
| "eval_f1": 0.7926421404682275, |
| "eval_loss": 0.7103803753852844, |
| "eval_runtime": 0.285, |
| "eval_samples_per_second": 1431.679, |
| "eval_steps_per_second": 7.018, |
| "step": 60 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 3.8317129611968994, |
| "learning_rate": 4.5e-05, |
| "loss": 0.2574, |
| "step": 75 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6936274509803921, |
| "eval_combined_score": 0.7373566326705814, |
| "eval_f1": 0.7810858143607706, |
| "eval_loss": 0.9412671327590942, |
| "eval_runtime": 0.2734, |
| "eval_samples_per_second": 1492.486, |
| "eval_steps_per_second": 7.316, |
| "step": 75 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 6.223960876464844, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.1479, |
| "step": 90 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6715686274509803, |
| "eval_combined_score": 0.7117102396514161, |
| "eval_f1": 0.7518518518518519, |
| "eval_loss": 1.1874147653579712, |
| "eval_runtime": 0.2693, |
| "eval_samples_per_second": 1515.172, |
| "eval_steps_per_second": 7.427, |
| "step": 90 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 5.585691928863525, |
| "learning_rate": 4.3e-05, |
| "loss": 0.0953, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6495098039215687, |
| "eval_combined_score": 0.6925922402232612, |
| "eval_f1": 0.7356746765249538, |
| "eval_loss": 1.4011175632476807, |
| "eval_runtime": 0.2886, |
| "eval_samples_per_second": 1413.947, |
| "eval_steps_per_second": 6.931, |
| "step": 105 |
| }, |
| { |
| "epoch": 7.0, |
| "step": 105, |
| "total_flos": 3377819728711680.0, |
| "train_loss": 0.3727283318837484, |
| "train_runtime": 60.2116, |
| "train_samples_per_second": 3045.927, |
| "train_steps_per_second": 12.456 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 750, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 50, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 5 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3377819728711680.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
